wtf-p 0.4.0 → 0.5.0

package/vendors/claude/commands/wtfp/new-paper.md

@@ -1,404 +1,169 @@
 ---
 name: wtfp:new-paper
-description: Initialize a new paper with deep context gathering and PROJECT.md
+description: Start a new paper with guided interview and setup
 allowed-tools:
   - Read
   - Bash
   - Write
+  - Glob
+  - Grep
+  - Task
   - AskUserQuestion
 ---
 
-<objective>
-
-Initialize a new academic paper through comprehensive context gathering.
-
-This is the most leveraged moment in any writing project. Deep questioning here means better structure, better arguments, better outcomes.
-
-Creates `.planning/` with PROJECT.md, config.json, and structure documents.
-
-</objective>
-
 <execution_context>
-
 @~/.claude/write-the-f-paper/references/principles.md
 @~/.claude/write-the-f-paper/references/questioning.md
 @~/.claude/write-the-f-paper/templates/project.md
 @~/.claude/write-the-f-paper/templates/config.json
 @~/.claude/write-the-f-paper/venues/
-
 </execution_context>
 
-<process>
-
-<step name="setup">
-
-**MANDATORY FIRST STEP — Execute these checks before ANY user interaction:**
-
-1. **Abort if project exists:**
-   ```bash
-   [ -f .planning/PROJECT.md ] && echo "ERROR: Paper already initialized. Use /wtfp:progress" && exit 1
-   ```
-
-2. **Initialize git repo in THIS directory** (required for version tracking):
-   ```bash
-   if [ -d .git ] || [ -f .git ]; then
-       echo "Git repo exists in current directory"
-   else
-       git init
-       echo "Initialized new git repo"
-   fi
-   ```
-
-3. **Detect existing writing materials (brownfield detection):**
-   ```bash
-   # Check for existing writing files
-   WRITING_FILES=$(find . -name "*.tex" -o -name "*.md" -o -name "*.bib" -o -name "*.docx" 2>/dev/null | grep -v node_modules | grep -v .git | grep -v .planning | head -20)
-   HAS_SOURCES=$([ -d sources ] || [ -d references ] || [ -d literature ] && echo "yes")
-   HAS_SOURCE_MAP=$([ -d .planning/sources ] && echo "yes")
-   ```
-
-   **You MUST run all bash commands above using the Bash tool before proceeding.**
+<objective>
+Initialize a new academic paper through comprehensive context gathering.
 
-</step>
+**Orchestrator role:** Check preconditions, detect brownfield projects, gather paper vision through batched questioning, synthesize PROJECT.md + config.json + structure documents, commit.
 
-<step name="brownfield_offer">
+Creates `.planning/` with PROJECT.md, config.json, and structure documents.
+</objective>
 
-**If existing writing detected and .planning/sources/ doesn't exist:**
+<context>
+No arguments. Runs in current directory.
+</context>
 
-Check the results from setup step:
-- If `WRITING_FILES` is non-empty OR `HAS_SOURCES` is "yes"
-- AND `HAS_SOURCE_MAP` is NOT "yes"
+<process>
 
-Use AskUserQuestion:
-- header: "Existing Material"
-- question: "I detected existing writing materials. Would you like to map them first?"
-- options:
-  - "Map sources first" — Run /wtfp:map-project to index existing literature/drafts (Recommended)
-  - "Skip mapping" — Proceed with paper initialization
+## 1. Validate Environment
 
-**If "Map sources first":**
-```
-Run `/wtfp:map-project` first, then return to `/wtfp:new-paper`
+```bash
+[ -f .planning/PROJECT.md ] && echo "ERROR: Paper already initialized. Use /wtfp:progress" && exit 1
 ```
-Exit command.
 
-**If "Skip mapping":** Continue to type step.
-
-**If no existing materials detected OR sources already mapped:** Continue to type step.
+Initialize git if needed:
+```bash
+[ -d .git ] || [ -f .git ] || git init
+```
 
-</step>
+## 2. Brownfield Detection
 
-<step name="initial_batch">
+```bash
+WRITING_FILES=$(find . -name "*.tex" -o -name "*.md" -o -name "*.bib" -o -name "*.docx" 2>/dev/null | grep -v node_modules | grep -v .git | grep -v .planning | head -20)
+HAS_SOURCES=$([ -d sources ] || [ -d references ] || [ -d literature ] && echo "yes")
+HAS_SOURCE_MAP=$([ -d .planning/sources ] && echo "yes")
+```
 
-**Gather Core Context (Batched):**
+**If materials detected and no source map:** Offer `/wtfp:map-project` first via AskUserQuestion.
 
-Use AskUserQuestion to collect the foundational pillars in one turn.
+## 3. Gather Core Context (Batched)
 
+Use AskUserQuestion — collect foundations in one turn:
 - header: "Paper Foundations"
-- question: "To initialize the project, I need to understand 4 things. Answer what you can, and I'll infer or ask about the rest:\n\n1. **Type:** (Research paper, Grant, Thesis, etc.)\n2. **Venue:** (Target conference/journal, e.g., NeurIPS, CHI, Nature)\n3. **Core Argument:** (The one main thing you are proving/proposing)\n4. **Audience:** (Who is this for?)"
-- options:
-  - "I provided the details" — Continue with what I said
-  - "Guide me step-by-step" — I prefer individual questions
-
-**If "Guide me step-by-step":**
-Fall back to the sequential interview mode (ask Type, then Venue, then Argument).
+- question: "1. **Type** (Research paper, Grant, Thesis)\n2. **Venue** (NeurIPS, CHI, Nature, etc.)\n3. **Core Argument** (one thing you're proving)\n4. **Audience** (who is this for?)"
+- options: "I provided the details" | "Guide me step-by-step"
 
-**If "I provided the details":**
-Parse the response to extract: `document_type`, `target_venue`, `core_argument`, `target_audience`.
+If step-by-step selected, ask Type → Venue → Argument sequentially.
 
-</step>
+## 4. Venue Template Selection
 
-<step name="venue_template">
-
-**Select venue template (if not inferred):**
-
-If `document_type` or `target_venue` is ambiguous, ask specifically for the template:
-
-Use AskUserQuestion:
+If ambiguous, ask via AskUserQuestion:
 - header: "Venue Structure"
-- question: "Based on your input, which structure fits best?"
-- options:
-  - "ACM CS" — Systems/Databases (SIGMOD, SOSP)
-  - "IEEE CS" — HPC/Systems (SC, IPDPS)
-  - "ML/AI" — NeurIPS, ICML, ICLR style
-  - "Nature/Science" — Life Sciences IMRaD
-  - "Grant" — Funding Proposal
-  - "Thesis" — Dissertation Chapter
-  - "Other" — Custom
-
-**Load venue template:**
-(Load corresponding YAML from `~/.claude/write-the-f-paper/venues/`)
-
-</step>
+- options: "ACM CS" | "IEEE CS" | "ML/AI" | "Nature/Science" | "Thesis"
 
-<step name="refine_argument">
+Load corresponding YAML from `~/.claude/write-the-f-paper/venues/`.
 
-**Sharpen the Core:**
+## 5. Sharpen Core Argument
 
-If `core_argument` was weak or missing in the batch, drill down now.
-
-Use AskUserQuestion:
+If core_argument was weak, drill down:
 - header: "Refine Core"
-- question: "Let's sharpen the core argument. If reviewers remember ONE thing, what is it?"
-- options:
-  - "The Method" — A new way of doing X
-  - "The Findings" — We discovered Y
-  - "The Theory" — A new perspective on Z
-  - "The System" — We built system Q
-  - "Write my own" — (User types input)
-
-</step>
+- question: "If reviewers remember ONE thing, what is it?"
+- options: "The Method" | "The Findings" | "The Theory" | "The System"
 
-<step name="constraints_batch">
+## 6. Gather Constraints
 
-**Gather Constraints & Boundaries (Batched):**
-
-Use AskUserQuestion:
 - header: "Scope & Limits"
-- question: "Finally, define the box we are playing in:\n\n1. **Out of Scope:** What are we explicitly NOT doing?\n2. **Hard Constraints:** Page limits, deadlines, data restrictions?"
-- options:
-  - "Provided details" — Save and continue
-  - "No constraints" — Standard length, no specific exclusions
-
-</step>
+- question: "1. **Out of Scope?** 2. **Hard Constraints?** (page limits, deadlines)"
+- options: "Provided details" | "No constraints"
 
-<step name="project">
+## 7. Synthesize PROJECT.md
 
-Synthesize all context into `.planning/PROJECT.md`:
+Write `.planning/PROJECT.md` using template from `~/.claude/write-the-f-paper/templates/project.md`.
 
-```markdown
-# [Paper Title]
+Populate: What This Is, Core Argument, Requirements (Must/Should/Out of Scope), Target Audience, Constraints, Key Decisions.
 
-## What This Is
-[Paper type, target venue, approximate length]
+## 8. Create Structure Documents
 
-## Core Argument
-[The thesis in 1-2 sentences — the one thing reviewers should remember]
+Create `.planning/structure/` with:
+- **argument-map.md** — Central thesis, supporting claims, logical flow, gaps
+- **outline.md** — Venue template sections, word budget table
+- **narrative-arc.md** — Problem/Journey/Resolution, reader experience
 
-## Requirements
+## 9. Workflow Mode & Config
 
-### Must Have
-- [ ] [Required element 1]
-- [ ] [Required element 2]
-- [ ] [Required element 3]
+Ask mode (Interactive / Flow) and depth (Quick / Standard / Comprehensive) via AskUserQuestion.
 
-### Should Have
-- [ ] [Desired element 1]
-- [ ] [Desired element 2]
+## 9b. Git & Planning Settings
 
-### Out of Scope
-- [Exclusion 1] — [why]
-- [Exclusion 2] — [why]
-
-## Target Audience
-[Who reads this, what they care about, what they already know]
-
-## Constraints
-- [Constraint 1]: [detail]
-- [Constraint 2]: [detail]
-
-## Key Decisions
-
-| Decision | Rationale | Outcome |
-|----------|-----------|---------|
-| [Choice from questioning] | [Why] | — Pending |
-
----
-*Last updated: [date] after initialization*
-```
-
-Do not compress. Capture everything gathered.
-
-</step>
-
-<step name="structure">
-
-**Create structure documents:**
-
-Create `.planning/structure/` directory.
-
-**argument-map.md:**
-```markdown
-# Argument Map
-
-## Central Thesis
-[From core argument above]
-
-## Supporting Claims
-### Claim 1: [Statement]
-- Evidence: [What will support this]
-- Strength: [to be determined]
-
-### Claim 2: [Statement]
-- Evidence: [What will support this]
-- Strength: [to be determined]
-
-## Logical Flow
-[How claims build to thesis]
-
-## Gaps
-- [Claims needing evidence]
-- [Logical leaps to address]
-```
-
-**outline.md:**
-```markdown
-# Document Outline
-
-## Venue Template: [acm-cs / ieee-cs / arxiv-ml / nature / thesis / custom]
-
-## Sections
-[Generated from venue template - see ~/.claude/write-the-f-paper/venues/]
-
-| # | Section | Purpose | Word Ratio |
-|---|---------|---------|------------|
-[Populated from venue template structure]
-
-## Word Budget
-| Section | Target | Current |
-|---------|--------|---------|
-| Abstract | [X] | 0 |
-| [Section 1 from template] | [X] | 0 |
-| [Section 2 from template] | [X] | 0 |
-| [...] | | |
-| Total | [X] | 0 |
-
-## Venue Notes
-[From venue template notes field - venue-specific guidance]
-```
-
-**narrative-arc.md:**
-```markdown
-# Narrative Arc
-
-## Act 1: The Problem
-- World state: [Current understanding]
-- Tension: [What's missing]
-- Stakes: [Why it matters]
-
-## Act 2: The Journey
-- Approach: [How you tackled it]
-- Discovery: [What you found]
-
-## Act 3: The Resolution
-- Insight: [What it means]
-- Transformation: [How understanding changes]
-
-## Reader Experience
-- Hook: [Why they keep reading]
-- Payoff: [The "aha" moment]
-```
-
-</step>
-
-<step name="mode">
-
-Ask workflow mode preference:
-
-Use AskUserQuestion:
-
-- header: "Mode"
-- question: "How do you want to work?"
+Ask about git tracking via AskUserQuestion:
+- header: "Git & Planning"
+- question: "Track planning documents in git?"
 - options:
-  - "Interactive" — Confirm at each step
-  - "Flow" — Auto-approve, minimize interruptions
-
-</step>
+  - "Yes (default)" -- commit_docs: true, planning artifacts are version-controlled
+  - "No" -- commit_docs: false, only paper content is committed
 
-<step name="depth">
+Use the answer to set `planning.commit_docs` in config.json.
 
-Ask planning depth preference:
+Ask about branching strategy via AskUserQuestion:
+- header: "Git Branching"
+- question: "Branch strategy for git history?"
+- options:
+  - "None (default)" -- all commits on current branch, git.branching_strategy: "none"
+  - "Section branches" -- one branch per section, merged on review pass, git.branching_strategy: "section"
+  - "Submission branches" -- branch per milestone/submission target, git.branching_strategy: "submission"
 
-Use AskUserQuestion:
+Use the answer to set `git.branching_strategy` in config.json.
 
-- header: "Depth"
-- question: "How thorough should section planning be?"
-- options:
-  - "Quick" — Draft fast, iterate later (minimal planning)
-  - "Standard" — Balanced planning and writing
-  - "Comprehensive" — Thorough outlines before writing
-
-Create `.planning/config.json`:
-```json
-{
-  "mode": "[chosen mode]",
-  "depth": "[chosen depth]",
-  "document_type": "[paper/grant/thesis/other]",
-  "venue_template": "[acm-cs/ieee-cs/arxiv-ml/nature/thesis/custom]",
-  "citation_style": "[from venue template]",
-  "gates": {
-    "confirm_project": true,
-    "confirm_sections": true,
-    "confirm_outline": true,
-    "confirm_plan": true,
-    "review_before_submit": true
-  }
-}
-```
+## 9c. Write Config
 
-</step>
+Write `.planning/config.json` with the full v0.5.0 schema: mode, depth, document_type, output_format, model_profile ("balanced"), venue_template, gates (8 gate types: confirm_outline, confirm_plan, confirm_write, confirm_review, execute_next_section, issues_review, confirm_transition, confirm_submission), writing, workflow, verification, planning (commit_docs, search_gitignored), parallelization, safety (always_confirm_destructive, always_confirm_external_services, backup_before_major_edits), git (branching_strategy, section_branch_template, submission_branch_template, squash_on_merge).
 
-<step name="commit">
+## 10. Commit
 
 ```bash
 git add .planning/PROJECT.md .planning/config.json .planning/structure/
-git commit -m "$(cat <<'EOF'
-docs: initialize [paper-title]
-
-[One-liner core argument]
-
-Creates PROJECT.md with paper vision and structure documents.
-EOF
-)"
+git commit -m "docs: initialize [paper-title]"
 ```
 
-</step>
-
-<step name="done">
+</process>
 
-Present completion with next steps:
+<offer_next>
 
-```
-Paper initialized:
+━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
+ WTF-P ► PAPER INITIALIZED ✓
+━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
 
 - Project: .planning/PROJECT.md
-- Config: .planning/config.json (mode: [chosen mode])
-- Structure: .planning/structure/ (argument-map, outline, narrative-arc)
-[If .planning/sources/ exists:] - Sources: .planning/sources/ (literature, data, prior-drafts)
+- Config: .planning/config.json (mode: [chosen])
+- Structure: .planning/structure/
 
----
+───────────────────────────────────────────
 
 ## ▶ Next Up
 
-**[Paper Title]** — create section outline
+**Create section outline**
 
 `/wtfp:create-outline`
 
 <sub>`/clear` first → fresh context window</sub>
 
----
-```
-
-</step>
+───────────────────────────────────────────
 
-</process>
-
-<output>
-
-- `.planning/PROJECT.md`
-- `.planning/config.json`
-- `.planning/structure/argument-map.md`
-- `.planning/structure/outline.md`
-- `.planning/structure/narrative-arc.md`
-
-</output>
+</offer_next>
 
 <success_criteria>
-
 - [ ] Deep questioning completed (not rushed)
 - [ ] PROJECT.md captures paper vision with core argument
 - [ ] Structure documents created (argument-map, outline, narrative-arc)
-- [ ] config.json has workflow mode
+- [ ] config.json has gates, safety, planning, and git sections
+- [ ] config.json has workflow mode and model_profile
 - [ ] All committed to git
-
 </success_criteria>

package/vendors/claude/commands/wtfp/pause-writing.md

@@ -1,6 +1,6 @@
 ---
 name: wtfp:pause-writing
-description: Create context handoff when pausing work mid-section
+description: Save current progress so you can resume later
 allowed-tools:
   - Read
   - Bash